Use when launching a long-running CLI in the background (external review CLIs, builds, batch jobs) or deciding how to capture its output; when a subagent is stalled, its completion is overdue, or its completion notification arrives while the work it launched…
Use when asking an external agent or subagent to review something or give a second opinion, applying review feedback, deciding whether to accept a finding, running two reviews at once, or before calling work "reviewed" and merging it. Triggers:…
Moomoo API (ex-Futu) Python SDK reference. Use for stock prices, orders, positions, or OpenQuoteContext/OpenTradeContext usage.
Common procedure and pitfalls for launching another coding agent's CLI (codex / claude / gemini) as a subprocess from the current session. Use before running `codex exec`, `claude -p`, or `gemini -p` — whether a review / second-opinion skill routes here, or…
Reviews Python code for type safety, async patterns, error handling, and common mistakes. Use when reviewing .py files, checking type hints, async/await usage, or exception handling.
user invoke only
Reviews and improves existing Claude skill files (SKILL.md) against Anthropic's best practices. Use this skill whenever a user shares a SKILL.md or skill folder, asks to "review", "refine", "improve", "audit", or "check" a skill, reports that a skill isn't…
user invoke only